Although meiosis has been observed for more than a century, many aspects of the process are still not well-understood. This is a very active field of research, so any summary of it will be incomplete and individualistic. Meiosis is studied by a variety of methods: genetics, biochemistry and molecular biology, light and fluorescence microscopy, and electron microscopy, as well as more recently by the tools of bioinformatics. One challenge is to connect the information gained from one type of experimental method with what is known from other methods. Two comprehensive reviews about the early stages of meiosis can be found at Zicker and Kleckner, 1998 and 1999, in the Annual Review of Genetics. Here are some of the key questions, as I understand them.
How do homologous chromosomes find each other and pair? This is one of the most actively researched fields. Some have suggested that homologues are found closer together than expected in pre-meiotic cells, as if the homologues are connected or (perhaps more likely) occupy a particular domain within the nucleus. Homologue recognition probably involves DNA sequence recognition, but no one has identified particular DNA sequences that are involved or proteins unique to homologue recognition. In both Drosophila and C. elegans, particular regions near the ends of chromosomes have been suggested as pairing regions—the part of the chromosome that pairs first before the rest of the chromosomes synapse. (The zipper is a very attractive analogy.) The candidate pairing regions have been found using genetic strains with large chromosome rearrangements (such as deletions, translocations, and duplications), which could be altering many properties of the chromosome as well as its sequence integrity. As appealing as this hypothesis is, conclusive proof has been difficult.
What is the role of the synaptonemal complex? This is one of the great mysteries of meiosis. Except for a few yeasts and fungi, all organisms that undergo meiosis have a elaborate structure known as the SC that forms between the homologues during prophase I. At the ultrastructural (i.e., electron microsopy) level, the SC of one species looks nearly identical to the SC of any other species. The distance between the homologues is the same (100 nm), the structure of axial elements and transverse elements looks the same, the
from Zicker and Kleckner, 1998
time of assembly is about the same. The major difference is in the length of the SC in different species, and this is one measure of chromosome compaction. Mutants in which the SC does not form have been identified in a number of different species. Naively, it would seem that it should be easy to deduce the role of the SC from these mutants. However, it is hard to know what in the primary defect and what are pleiotropic effects. At least some mutants that do not assemble an SC still have reasonable levels of crossing over, so it is clear that the presence of an SC is not an absolute requirement for recombination. However, these mutants do not have interference. It is also true that S. pombe and Aspergillus nidulans, which do not have SC, also do not have interference. One appealing idea is that the SC is involved in interference. However, since we do not understand the role of the SC and we do not understand the mechanism of interference, we may simply be explaining one unknown by another unknown.
What is interference? Crossovers do not occur at random in the genome. Although it is generally true that genes that are further apart in physical distance will cross over more frequently than those that are close together, the rate of recombination per physical distance (e.g, the number of crossovers per megabase) varies widely. In addition, the presence of crossovers are not independent of each other, and the presence of one crossover reduces the probability of a second crossover nearby. This is known as interference. C. elegans is an extreme example of this since there appears to be only one crossover per chromosome (that is, interference is 100%). This means that a crossover in one region can prevent the formation of a crossover more than 15 megabases away. [Of course, “15 megabases away” reflects a linear arrangement of genes, and may not reflect the proximity of genes in the highly folded meiotic chromosome.] The molecular basis of interference is obscure.